Many Iranian metropolises, including Shiraz, are situated in arid and semi-arid regions, lacking sufficient renewable water resources. In recent years, climate changes, including drought and rising temperatures, have led to changes in water supply and demand. Given the necessity and importance of urban water supply, this study investigates the impact of different climate scenarios on residential water demand. Many studies, in their models, do not consider the social interactions between household water consumers and the change in their consumption behaviour, which serves as a fundamental drawback. Thus, the present research attempts to propose an agentbased framework by modelling social interactions via the diffusion process to investigate water consumption behaviour efficiently. The model is calibrated and applied to Shiraz City in Iran, according to the data from 2006 to 2019, and it is used to simulate each scenario for the following years until 2032. The findings show that temperature has a positive and significant effect on residential water consumption; yet, rainfall negatively affects water consumption. The simulation results of these scenarios for temperature increase or decrease and rainfall changes are estimated. In addition, the developed agent-based platform can be easily calibrated and adjusted based on the data of any other city to simulate water demand estimation under different climatic and even economic scenarios. Urban water managers can benefit from such estimates to plan future infrastructure development and proactive management of seasonal water resources under the growing pressure of potential climate change because construing the sensitivity of seasonal water consumption to climate conditions is essential to respond to variations in demand.

Keywords